Munich, Germany – Wacker Chemie AG and CordenPharma International GmbH, together with Ludwig Maximilian University of Munich (LMU) and Humboldt University Berlin (HU Berlin), have launched a project to accelerate the development of RNA-based drugs. The aim is to develop a new generation of lipid nanoparticles (LNPs), which are an essential component of RNA-based drugs. Based on these formulations, a machine learning algorithm is to be trained that automatically identifies the ideal components for new RNA formulations – so far a particularly time-consuming and cost-intensive development step. The project, which will start on 1 April 2023, is scheduled to run for three years and is funded by the Federal Ministry for Economic Affairs and Energy with around € 1.4 million.
Following the success of RNA-based COVID-19 vaccines, drugs containing RNA as an active ingredient are believed to have great medical potential. The focus of development is not only on vaccines against infectious diseases, but also therapies against cancer and hereditary diseases. Various active ingredients with different lipid-nanoparticle compositions are being worked on worldwide. With their joint project, WACKER, CordenPharma, LMU and HU Berlin have set themselves the goal of accelerating the development of RNA-based drugs. To achieve this, the partners are developing a new generation of lipid nanoparticles (LNPs) as well as a machine learning system for RNA formulation that is expected to shorten development time and reduce costs.
The partners take on different tasks in the project. With the production of RNA molecules, WACKER supplies the core of RNA-based drugs. In addition to messenger ribonucleic acid (mRNA), which is the focus of clinical application, WACKER is also producing other RNA molecules for the project, such as self-amplifying RNAs (saRNA) and circular RNAs (circRNA). For these, the company tests new manufacturing processes. “The RNA molecule species have different properties, are suitable for different purposes and are produced differently,” explains Dr. Hagen Richter, head of nucleic acid research at WACKER and responsible for coordinating the funding project. “saRNA and mRNA are currently mainly used in vaccine development. circRNAs are characterized by higher stability and are therefore particularly suitable for therapies in which drugs have to be released more slowly and longer.” In recent years, WACKER has already built up expertise in the production of mRNA in accordance with GMP (Good Manufacturing Practice) guidelines.
In the project, CordenPharma will develop basic building blocks for nanoparticles, so-called modified lipids, together with HU Berlin. They ensure that the active ingredients enter the body safely and are released at the destination. “The development of lipid nanoparticles for RNA formulation is a complex process that requires special lipids. So far, LNP optimization is mainly based on screening of functional lipids in many time- and cost-intensive experiments. Machine learning, which is a branch of artificial intelligence, will help to understand the relationship between functional lipids and effective mRNA vaccines in cell culture experiments. This allows us to develop a new generation of lipids with improved properties that result in even more effective active ingredients,” says Dr. Adriano Indolese, Global Head of Development & Innovation at CordenPharma International. CordenPharma and HU Berlin will synthesize the new lipid components and analyze them physicochemically in conjunction with the various RNA molecules. The cellular functionality of the formulations is then investigated at LMU in cell culture experiments. This shows how targeted and well the active ingredients are released. The screening of different RNA types in conjunction with the modified lipids is intended to create the broadest possible database.
The data from the physical, chemical and biological analyses of the LNPs and the various RNA molecules are used to train a machine learning algorithm for RNA formulations. Machine learning is about artificial intelligence learning learning from examples and being able to generalize them after the end of the learning phase. Specifically, the system, which will be set up at LMU, will use the properties of the LNPs to provide precise allocation to various RNA molecules and ultimately forms of therapy. After the learning phase, the algorithm should be able to assign suitable formulation approaches to any RNA molecule. After training the system, the functionality of the system will be tested in a concrete application in the last phase of the three-year project. The Federal Ministry for Economic Affairs and Energy is funding the joint project with around € 1.4 million.